Fuel cycle studies on minor actinide transmutation in Generation IV fast reactors

Transcription

1 Fuel cycle studies on minor actinide transmutation in Generation IV fast reactors M. Halász, M. Szieberth, S. Fehér Budapest University of Technology and Economics, Institute of Nuclear Techniques

2 Contents Introduction The FITXS burn-up method GFR-LWR and LFR-LWR fuel cycle studies The Gas-cooled Fast Reactor (GFR) and Lead-cooled Fast reactor (LFR) models GFR-ER symbiotic fuel cycle studies The European ressurized Reactor (ER) model Analysis of the neutron economies Conclusions 2

3 Introduction Fast spectrum reactors: The average neutron yield per fission increases with energy σ f /σ a increases with energy Higher uranium utilization ratio and more favourable waste composition Role of fuel cycle simulation codes: Assistance in strategic decisions Classification of waste: isotopic composition, decay heat, radiotoxicity Infrastructural issues: reprocessing capacity, storages, etc. Spent fuel composition of the reactors has to be calculated 3

4 Determination of the spent fuel composition The equation system which describes the evolution of the fuel composition: d N( t) dt = AN( t) ( λ σφ) N( t) Difficulty: σ σ( N( t)), Φ Φ( N( t)) The neutron spectrum changes with the isotopic composition therefore influencing the one-group averaged cross-sections and the integrated neutron flux. Usual approximations: Burn-up tables Cross-section libraries for specific burn-up arametrization of cross-sections 4

5 The FITXS method Idea: parametrization of the one-group averaged cross-sections as functions of the detailed fuel composition Selection of fitting parameters reparation of cross-section database Cross-section + k eff fitting Selection of fitting parameters: actinides with significant contribution to the reaction rates + overall quantity of fission products parameters reparation of cross-section database: numerous (few thousand) core calculations with randomly sampled fuel compositions Cross-section + k eff fitting: σ σ( N 1,..., Nn) a 0 j a j N j j, k a jk N j N k M. Szieberth, M. Halász, S. Fehér, T. Reiss, Simulation of fuel cycles with minor actinide management using a fast burnup calculation tool in: roc. of HYSOR 2014, Kyoto, apan, 28 September 3 October 2014 M. Szieberth, M. Halász, T. Reiss, S. Fehér: Fuel cycle studies on minor actinide burning in gas cooled fast reactors" in: roc. of Fast Reactors and Related Fuel Cycles, aris, France, March 2013 (FR13) 5

6 The Gas-cooled Fast Reactor (GFR) model 3D SCALE model of the GFR2400 core (T. Reiss) Homogenized inner and outer core fuel assemblies with different u content Random compositions with the following constraints: Homogeneous MA and F load u inner /u outer = 0,8 0-10% MA content, 10-25% u content, the rest is U Fission products with average relative composition Isotopic compositions of actinide elements are also randomly sampled 6

7 The Lead-cooled Fast Reactor (LFR) model reliminary calculations for the analysis of the ELSY (European Lead-cooled SYstem) core 3D SCALE fuel assembly model in infinite lattice Random compositions with the following constraints: Average fuel composition 0-10% MA content, 13-22% u content, the rest is U Fission products with average relative composition Isotopic compositions of actinide elements are also randomly sampled 7

8 Fitting results The fitted polynomials can describe the cross-sections and k eff with an average error well below 1% 239 u (n,f) cross-section (GFR) 8

9 Fitting results The fitting errors have mean values close to zero, and are related to the statistical uncertainties of the Monte Carlo transport calculations 239 u (n,f) cross-section (GFR) Fitting errors vs. statistical uncertainties 9

10 Verification of the burn-up models Burn-up calculations using the fitted cross-sections are in excellent agreement with SCALE based burn-up results with less than 0,1% error for main isotopes 10

11 GFR-LWR and LFR-LWR fuel cycle studies The developed burn-up models were integrated into a fuel cycle models containing fast reactors and conventional Light Water Reactors (LWRs) 11

12 GFR and LFR closed fuel cycle results GFR End-of-Cycle u content u feed from LWRs No external u feed is needed in the equilibrium GFR reaches self-breeding The u content of the core increases to counterweight the decrease of fissile u 12

13 GFR and LFR closed fuel cycle results LFR End-of-Cycle u content u feed from LWRs The LFR also reaches breakeven breeding Both reactors are able to breed their fuel from fertile 238 U 13

14 GFR and LFR closed fuel cycle results GFR End-of-Cycle MA content LFR End-of-Cycle MA content Both the GFR and LFR reaches an equilibrium MA content of 1% with no Cm accummulation Every minor actinide isotope can be burned in the hard neutron spectrum of the GFR and LFR 14

15 GFR and LFR closed fuel cycle results GFR MA consumption LFR MA consumption The GFR and LFR are able to burn additional minor actinides from LWRs if more MAs are fed to the core than the equilibrium MA content 15

16 GFR and LFR closed fuel cycle results Minor actinide waste due to reprocessing losses (GFR left, LFR right) The MA burning capabilities of the reactors highly exceed the MA waste production from partitioning losses Nevertheless, the partitioning technology has main impact on the final waste 16

17 GFR and LFR closed fuel cycle results GFR breeding gain LFR breeding gain Higher MA content of the GFR and LFR cores results in higher breeding gain This effect can be explained by the analysis of the neutron economy 17

18 The European ressurized Reactor (ER) model Generation III Light Water Reactor Goal: investigation of the recycling of excess u produced by the FRs in Light Water Reactors 3D SCALE fuel assembly model in infinite lattice Random compositions with the following constraints: Average fuel composition 5-10% u content, 0-1% MA content, the rest is U Fission product composition fitted as function of assembly burnup + Xe, Sm calculated explicitly Isotopic compositions of actinide elements are also randomly sampled 18

19 GFR-ER symbiotic fuel cycle studies The u produced by the GFR is recycled in ER MOX fuel assemblies, and the spent MOX u is recycled in the GFR 19

20 GFR-ER symbiotic fuel cycle results GFR End-of-Cycle MA content GFR End-of-Cycle u composition The GFR is able to burn the minor actinides from the spent MOX fuel, but the recycling of the excess u results in high MA content and lower fissile u ratio Transmutational capabilities and possibly safety parameters degraded 20

21 Analysis of the neutron economies Motivation: deeper understanding of fuel breeding and transmutation capabilities D-value = neutron consumption / fission (M. Salvatores, 1993) where R D i 1 { R i 1 i 1i 2 [ R k 1 i 1i 3 (...)]} n 2 3 (n, γ), decay, (n,f), (n,2n). Neutron consumption of the fuel: k 1 i leakage 1 i n, a f 1 construction materials i breeding / transmutation D FUEL D S D ( L CM ) ext M. Salvatores et al: A Global hysics Approach to Transmutation of Radioactive Nuclei, Nuclear Science and Engineering Vol. 116, 1994, pp FUEL D A 21

22 22 Analysis of the neutron economies The evaluation of the D-values requires the consideration of every possible transmutational chains, making the calculation difficult k n n i i k i i i i R R D (...)]} [ {

23 Analysis of the neutron economies Based on the theory of Markov-chains we derived a closed formula for the neutron consumption / production Transition probability matrix: ( t0, t1 ) ( t) xy xy ( t ) xy, (0) I The neutron production can be expressed as one step plus the neutron production of the next nuclide in the chain: n D x ~ ~ xy xy 1 2 n n D xy D xy 1 2,, xy xy 1 2 probability probability xy xy x, a f xy x x Tower rule + conditional expectation value: ~ ˆ D n D v ~ xx 1, ~ xy xy, v D x xf ~ xf y R xy ~ xy 23

24 Analysis of the neutron economies Neutron production of actinide isotopes in the two examined fast reactor types Isotope Neutron production / fission GFR LFR 235 U U Np u u u u u Am m Am Am Cm Cm Cm

25 Analysis of the neutron economies Neutron production of actinide isotopes with respect to number of transitions MA isotopes are fertile if not fissile in the hard neutron spectrum of the GFR High MA content increases the total neutron production therefore improves breeding capabilities 25

26 Analysis of the neutron economies Time-dependent neutron productions based on continuous-time Markov-chain 237 Np 243 Am The time-dependency of the average neutron productions also shows the fertile nature of otherwise absorbing minor actinides 26

27 The SITON code The method has been also implemented in the SITON nuclear fuel cycle simulation code developed in the Hungarian National Academy of Sciences. SITON v2.0 is a dynamic, discrete facilities/discrete materials and also discrete events fuel cycle simulation code. Also, SITON v2.0 is involved in the Benchmark Study on Nuclear Fuel Cycle Transition Scenarios Analysis Codes organized by the OECD Working arty on Scientific Issues of the Fuel Cycle (WFC) Á. Brolly et al.: hysical model of the nuclear fuel cycle simulation code SITON, in Annals of Nuclear Energy (2016) 27

28 Summary GFR, LFR and ER MOX burn-up models were developed with the FITXS method GFR-LWR and LFR-LWR fuel cycles were investigated with and without MOX recycling in ER MOX fuel assemblies Both the GFR and the LFR is able to operate with u from LWR spent fuel An equilibrium MA content of 1% is reached during closed cycle operation, and both FRs are capable of burning additional MAs from LWR spent fuel MOX u recycling in ERs results in significant deterioration of the u isotopic composition, as well as higher equilibrium MA content Higher MA content results in higher breeding gain due to fertile nature of MA isotopes in fast neutron spectrum Although MOX recycling increases uranium utilization ratio compared to FR-LWR fuel cycle, higher MA content and worse u composition suggest that recycling of excess u would be more favourable in FRs 28

29 Thank you for your attention! 29